Recently, many image files are being generated by a digital camera, a scanner and etc. and they are widely distributed via Internet. However, a slew of images are corrected by an image editing program such as a photo shop, and a problem was generated that it is not possible to distinguish whether a digitized image file is an original one which is initially generated in an actual device or a corrected file.
In the past, in case of an electronic document, there was an system, wherein the code values of the characters described in the document are embedded into two-dimensional bar codes, and then distributed, and forgery of a document can be verified by comparing the code values of the characters of the distributed document and the two-dimensional bar codes.
But, in case of an image file, since the size of data became large when embedding all pixel information into a two-dimensional bar code, it is impossible to do so. For example, a digital camera which is used very frequently recently has approximately 5 million pixels, but the data amount including RGB color value becomes approximately 15 mega bites. Therefore, there is a shortcoming that even two-dimensional bar code which can contain tremendous amount of data can not contain all pixel information of an image file.